Well, if you think of it as some generalized instance of the secretary problem, the 40-70% rule wouldn't necessarily be that bad --I guess it depends on how adv

Well, if you think of it as some generalized instance of the secretary problem, the 40-70% rule wouldn't necessarily be that bad --I guess it depends on how adversarial are your assumptions [1]. [1] http://math.stackexchange.com/questions/33227/3-other-extensions-of-the-secretary-problem http://math.stackexchange.com/questions/33227/3-other-extens...